Watt is a volatility farming protocol on Solana that uses natural market arbitrage to generate real yield for users and foster liquidity for other projects in the ecosystem.
No related events yet.
Performance data for the previously reported OpenAI in-house inference chip Jalapeño was publicly disclosed for the first time today. In tests across the GPT-OSS 120B, DeepSeek R1 670B, and Kimi K2.5 1T models, Jalapeño's AI throughput per watt was 1.5 to 1.9 times higher than that of the GB200 and GB300, reaching 1.7 times that of the GB300 specifically under the DeepSeek R1 workload. Its end-to-end latency is approximately 28% to 59% of that of the compared systems.
OpenAI researcher Jeffrey Wang described the performance of the AI inference solution jointly developed by AMD and Cerebras as "incredible". The solution integrates the AMD Helios rack with the Cerebras wafer-scale engine, handling prompts and token generation stages separately, and is estimated to increase tokens per second per watt by 5 times. This technical approach optimizes inference performance through heterogeneous computing, reflecting the continued focus of leading AI labs on reducing inference costs.
Anza CEO Brennan Watt stated that Solana-related SIMD proposals are expected to be finalized this year. Among them, SIMD-123 has already been approved and is nearing code completion; the discussion draft for SIMD-547 is substantively aligned with SIMD-553, and both SIMD-553 and SIMD-550 have received concept ACKs from Anza. If SIMD-550 and SIMD-553 are implemented jointly, the annual SOL inflation reduction rate would increase from 15% to 30%, reducing token emissions by approximately $1.36 billion over six years under current price assumptions. Additionally, daily SOL burn volume would rise from roughly 650 tokens (approximately $47,000) to a maximum of approximately 9,000 tokens (approximately $646,000).